Oct. 14. 1918 Fertilizer Potash Residues in Hagerstown Silty Loam 61 



Moreover, in five groups of plots from the same field, each with two 

 plots receiving no potash and two that were periodically dressed with 

 potash fertilizers, the soils of the potash-dressed pair invariably con- 

 tained more total potash than those of the nonpotash pair; with a single 

 exception, more potash soluble in hydrochloric acid; and in all cases 

 much more potash soluble in 1 per cent citric acid, the average ratio for 

 the nonpotash to. potash-treated soils in this last case averaging i to 9. 



PENNSYLVANIA STUDIES 



The soils used for the present study represent plots i and 4, Tier II, 

 of the General Fertilizer Experiments of the Pennsylvania Experiment 

 Station, which have been maintained continuously since 1881. All the 

 plots of Tier II were cultivated and cropped alike, the land being kept 

 under the common Pennsylvania 4-course rotation of corn {Zea mays), 

 oats (Avena saliva), wheat (Triticum aestivum), and grass (mixed timothy 

 {Phleum pratense) and medium red clover (Trifolium pratense). Plot i 

 received no dressings of any kind during this experiment. Plot 4 was 

 dressed biennially, for corn and wheat, with 2o<3 pounds of muriate of 

 potash, so that, beginning with 1881, it had received 18 such dressings, 

 equivalent to i ,800 pounds of potash (K^O) up to the time of the sampling 

 in 1 91 6, when the land was in oat stubble. The latest of these dressings 

 had been applied 14 months prior to the sampling. 



From 1868 to 1881 the plots were used for certain cultivation experi- 

 ments, plot I being plowed with a common plow for corn and wheat, 

 plot 4 with a subsoil plow for corn, and a Michigan plow for wheat. 

 Both plots were, however, cropped alike and were also fertilized alike 

 and not at all heavily. Prior to 1868 the land was under general farm 

 cultivation and cropping, the two plots forming part of the same field 

 and doubtless ha,\nng had the same farm history. 



The weight of the acre 7-inch surface layers of the two plots was 

 determined ^ at 22 points uniformly distributed over each plot, with de- 

 tails of method and result that have been elsewhere reported.^ The air- 

 dry (not oven-dry) weights of the respective surface 7-inch layers were: 

 Plot I, 2,091,662 pounds to the acre; plot 4, 2,036,449 pounds to the 

 acre. The probable error of these determinations was less than 10,000 

 pounds to the acre. 



The 22 subsamples obtained in two independent series from each plot 

 in the course of the acre-weight determinations were supplemented by 

 80 other subsamples, likewise divided into two independent series for 

 each plot, obtained by means of a soil auger. Both sets of subsamples 

 were prepared in the same manner, by air-drying and sifting all of each 



1 All quantitative determinations reported were, unless otherwise specifically credited, made by Mr. 

 Erb under the supervision of the senior author. 



2 Frear. William, and Erb. E. S. excavation method for determinwg the apparent speofic 

 GRAVITY OF SOILS. To be published in Proc. Assoc. Off. Agr. Chem. 1917. 



